Rules at a Glance

The rules every submission must follow, collected in one place. Each rule has a stable ID you can link to (for example rules-at-a-glance/#R-08) or quote to the helpdesk. Error IDs (E-xx) point to Troubleshooting Errors.

All Packages Templates IDs & references Values Privacy Access & changes

ID	Rule	Details
R-01	One study per package.	The first study accession seen fixes the study for the whole package. Data for a second study fails (E-29).
R-02	Web uploads are limited to 512 MB per package.	For larger packages use IBM Aspera, or the API's offline or S3 uploads (API).
R-03	A package is a plain .zip.	No password, no split archives, not a renamed .rar or .7z (E-26).
R-04	Use the current template version (3.37) and keep its fixed rows intact.	Don't edit the template name / Schema Version row, the column header row or section headers (E-14, E-22). The only exception is updating the version number of unchanged templates, as described in the version history.
R-05	Keep every column in the header row, even ones you leave empty.	A removed column fails even if it's optional (E-23).
R-06	Each template appears once per package, in one format: .xlsx, tab-delimited .txt or .json.	No legacy .xls (E-19). No two files whose names differ only by case (E-24). Keep the standard template file names, without browser suffixes such as (1).
R-07	Required sections and columns must be filled in.	Required fields are bold in the Excel template. Conditionally required fields are not bold; check the trigger on each field page (E-07, E-09).
R-08	Every reference must resolve: the record must be in the same package or already in the workspace.	Spell IDs exactly the same in both places (E-18). See Upload Order.
R-09	Upload in dependency order across packages.	Study, then subjects, then clinical data and assays. Inside one package ImmPort sorts the templates itself (loading order).
R-10	For each assay, treatments, reagents and experiment samples go in the same package, together with the results template or results file.	See the dashed groups in Assay Workflows.
R-11	User-defined IDs are unique within the workspace and are never reused.	Including across studies and earlier uploads (E-04, E-05). For human subjects the ID must not be identifying.
R-12	Dates are DD-MMM-YYYY (or DD-MMM-YY), for example 06-Feb-2005.	ISO and numeric dates are rejected. Set Excel date columns to Text (E-20). Subject time points use study days instead of dates (R-19).
R-13	Controlled vocabulary fields accept only listed values.	Matching ignores case, and the stored value takes the list's spelling (E-16). Preferred vocabulary fields accept free text and map it when they can (E-17).
R-14	Numbers are bare numbers: no units, no thousands separators, the right kind (integer, positive, …).	(E-03)
R-15	Paired value and unit fields are filled together or left empty together.	(E-21)
R-16	Text must fit the column's maximum length.	Shown as Text, max N characters on each field page (E-01).
R-17	basic_study_design needs at least one row in planned_visit, inclusion_exclusion and study_2_protocol.	Planned visit and inclusion/exclusion accessions from another study can't be reused; protocol accessions can (E-11 to E-13).
R-18	subjectHumans and subjectAnimals only create new subjects.	To reuse an existing subject in another study, use study_design_edit → arm_2_subject in the same workspace. A subject belongs to one arm per study (how, E-28).
R-19	De-identify human subject data.	No direct identifiers, no subject calendar dates (use study days from day 0), no location below state level. Enter true ages; the uploader rounds ages over 89 to 90 (details).
R-20	Uploading needs the Submitter or Administrator role and write access to the study.	Viewers can't upload, through the website or the API (roles).
R-21	To change loaded data, contact the helpdesk, except for study design details.	Update study design details yourself with study_design_edit.
R-22	Validate before you upload.	The Validator runs a test load without saving anything. Fix the first errors first (how).
R-23	Every file named in a template is in the package.	Protocol documents, raw data and results files (E-02).
R-24	Either the ImmPort results template or your own results file, never both.	Controlled by the ImmPort Template? column of HLA, KIR, MBAA, mass spectrometry and RNA-seq experiment samples (E-31).